# Notes and Questions about SSZ Query Language


Yesterday, I sent [my initial draft](https://hackmd.io/@junsong/SJGze5cNxg) via Prysm's Discord server, and Bastin suggested to take a glimpse of [SSZ Query Language](https://github.com/eth-protocol-fellows/cohort-six/blob/master/projects/project-ideas.md#prysm-ssz-query-language) project as well. As it is tightly coupled to my initial interest ([Merkle Proofs of Everything](https://github.com/eth-protocol-fellows/cohort-six/blob/master/projects/project-ideas.md#prysm-merkle-proofs-of-everything)), I took a look of it. This post includes some notes and questions to the team.

## Notes

{
    "anchor": "<stateRoot>",
    "querySpec": {
        "validators": {
            "range": [
                0,
                100
            ],
            "filter": {
                "slashed": true,
                "effective_balance": {
                    "gte": 32
                }
            },
            "fields": [
                "pubkey",
                "effective_balance"
            ]
        }
    },
    "includeProof": true
}

Project [description](https://github.com/eth-protocol-fellows/cohort-six/blob/master/projects/project-ideas.md#prysm-ssz-query-language) contains an example query so that I can think of (or at least imagine) which features it should provide.

### Details

1. Without `includeProof` option, it's [a GraphQL approach](https://discord.com/channels/595666850260713488/804019759934078987/1328334567391887432) for SSZ as @fjl said.
2. To provide the feature for `includeProof`, [Merkle Proofs of Everything](https://github.com/eth-protocol-fellows/cohort-six/blob/master/projects/project-ideas.md#prysm-merkle-proofs-of-everything) should be completed. It MUST provide a proof for any path. (That's why Bastin says to look this project as well.)
3. As it should support multiple fields, it definitely needs to leverage what [Merkle Multiproof](https://github.com/ethereum/consensus-specs/blob/v1.5.0/ssz/merkle-proofs.md#merkle-multiproofs) does.
4. User will include his/her query in the body of HTTP POST request.

### An example for the Response

For the given query above, my very first draft for the response is following:

{
    "dependent_root": "<stateRoot>",
    "execution_optimistic": false,
    "data": {
        "validators": [
            {
                "index": 42,
                "fields": {
                    "pubkey": "0x93247f2209abcacf57b75a51dafae777f9dd38bc7053d1af526f220a7489a6d3a2753e5f3e8b1cfe39b56f43611df74a",
                    "effective_balance": "32000000000",
                }
            }
        ]
    },
    "proofs": {
        "leaves": [
            "0x...ssz_hash_of_the_pubkey_field...",
            "0x...ssz_hash_of_the_effective_balance...",
        ],
        "witnesses": [
            {
                "generalized_index": 1484895,
                "witness":  "0x...ssz_hash_of_the_withdrawal_credentials_field..."
            }, // and so on...
        ],
        "generalized_indices": [
            1484894, // Generalized Index for 42th validator's pubkey.
            1484898, // Generalized Index for 42th validator's effective_balance.
        ]
    },
}

To note, generalized indices are very important to verify the multiproof. So all data is coupled with generalized indices.

## Questions

1. Please correct me if I'm wrong or getting in the different way :)
2. It seems this proposal was suggested quite long time ago. I digged some EIPs that might deal with this feature, but I couldn't find the one. (The only material that I can find is a couple of paragraphs from Etan's post.) **Will this feature be proposed as an EIP, or just remain as a client feature?** Also, was there any bottleneck for implementing this feature for other client team and Prysm as well?
3. I think I can take the both projects up, and is there any priority between those projects? So that I might adjust the timeline for the next five months.

Thanks in advance!